Engineering an Alcohol Dehydrogenase for Balancing Kinetics in

Subscriber access provided by Georgetown University | Lauinger and Blommer Libraries

Article

Engineering an Alcohol Dehydrogenase for Balancing Kinetics in NADPH Regeneration with 1,4-Butanediol as a Cosubstrate Guo-Chao Xu, Cheng Zhu, Aitao Li, Yan Ni, Rui-Zhi Han, Jieyu Zhou, and Ye Ni ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.9b03879 • Publication Date (Web): 19 Aug 2019 Downloaded from pubs.acs.org on August 26, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

1

Engineering an Alcohol Dehydrogenase for

2

Balancing Kinetics in NADPH Regeneration with

3

1,4-Butanediol as a Cosubstrate

4

Guochao Xu†,¶, Cheng Zhu†,¶, Aitao Li‡, Yan Ni§, Ruizhi Han†, Jieyu Zhou†, and Ye Ni*,†

5

†Key

6

Jiangnan University, Wuxi 214122, Jiangsu, China.

7

‡Hubei

8

Laboratory of Industrial Biotechnology, College of Life Sciences, Hubei University, Wuhan

9

430062, China.

Laboratory of Industrial Biotechnology, Ministry of Education, School of Biotechnology,

Collaborative Innovation Center for Green Transformation of Bio-resources, Hubei Key

10

§Department

11

Eindhoven, Netherlands.

12

*Corresponding

13

Guochao Xu: [email protected];

14

Cheng Zhu: [email protected];

15

Aitao Li: [email protected];

16

Yan Ni: [email protected];

17

Ruizhi Han: [email protected];

18

Jieyu Zhou: [email protected];

19

Ye Ni: [email protected]

of Biomedical Engineering, Eindhoven University of Technology, NL-5600 MB

author: Prof. Y Ni. [email protected].

1 Environment ACS Paragon Plus

ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

1

ABSTRACT

2

Cofactor regeneration using diols as ‘smart cosubstrates’ is one of the most promising

3

approaches, due to the thermodynamic preference and 0.5-equivalent requirement. In order to

4

establish an efficient NADPH regeneration system with 1,4-butanediol (1,4-BD), a NADP+-

5

dependent alcohol dehydrogenase from Kluyveromyces polysporus (KpADH) was engineered to

6

solve the kinetic imbalance. Several hotspots were identified through molecular dynamic

7

simulation and subjected to saturation and combinatorial mutagenesis. Variant KpADHV84I/Y127M

8

exhibited a lower KM of 15.1 mM and a higher kcat of 30.1 min–1 than WTKpADH. The oxidation

9

of 1,4-BD to 4-hydroxybutanal was found to be the rate-limiting step, for which the kcat/KM value

10

of double mutant KpADHV84I/Y127M was 2.00 min–1·mM–1, 11.6-fold higher than that of

11

WTKpADH.

12

kcat/KM

13

KpADHV84I/Y127M was dramatically reduced by almost 100-fold compared to

14

was advantageous for NADPH regeneration. As much as 100 mM phenylpyruvic acid could be

15

reduced into D-phenylalanine with 99.2% conversion in 6 h using merely 0.5-equivalents of 1,4-

16

BD. Both the improved catalytic efficiency toward 1,4-BD and the balanced kcat/KM between 1,4-

17

BD and 2-HTHF contributed to the higher NADPH regeneration efficiency. This study provides

18

guidance for engineering alcohol dehydrogenases for cosubstrate specificity toward diols and its

19

application in NADPH regeneration for the preparation of chiral compounds of pharmaceutical

20

relevance.

21

KEYWORDS: NADPH, smart cosubstrate, 1,4-butanediol, cofactor regeneration, balancing

22

kinetics

KpADHV84I/Y127M preferred diols with longer chain length (C5-C6). The ratio of

toward

2-hydroxytetrahydrofuran

(2-HTHF),

in


comparison

to

1,4-BD,

WTKpADH,

in

which

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

2

INTRODUCTION

3

The application of biocatalysis in the preparation of chiral compounds has undergone a

4

revolution over the past decades from “the state of the art” to “the first choice for chemists”. The

5

desirability of biocatalysts stems from their outstanding characteristics, such as high

6

enantioselectivity, renewability, and biodegradability, qualities especially attractive to the

7

burgeoning field of green chemistry.1–3 This is particularly true for redox reactions, including

8

alcohol or amino acid dehydrogenases used in the reduction of the C=O bond, ene or enoate

9

reductases used for reduction of the C=C bond, and oxygenases for oxyfunctionalisation.4–6 The

10

chiral compounds produced are of high economic value and are used widely as food ingredients

11

and building blocks for the production of pharmaceuticals.7,8 Most of the reactions mentioned

12

above, however, require stoichiometric amounts of NAD(P)H, which can be very expensive.9 As

13

an economic and practical approach, NAD(P)H should be added in catalytic amounts and

14

regenerated in situ.10

15

Intensive research has been committed to the development of efficient NAD(P)H

16

regeneration systems, which have been accomplished by enzyme-coupled, substrate-coupled,

17

electro-enzymatic, photochemical, and chemical approaches.11–13 For a practical cofactor

18

regeneration approach, the cosubstrates should be inexpensive, the reactions should be

19

thermodynamically and kinetically favorable, and the coproducts should be separable and non-

20

inhibitory.14 Considering all these factors, well-established enzyme- and substrate-coupled

21

approaches are currently more attractive for regeneration of nicotinamide cofactors.8 Alcohol

22

dehydrogenase (ADH)-catalyzed oxidation of simple alcohols (such as ethanol and isopropanol)



Page 4 of 36

1

represents the most elegant example of this approach (Table 1). Ethanol is a four-electron

2

reductant, and it is especially attractive as a cosubstrate due to its favorable redox potential FT6D

3

of –7.4 kcal·mol–1), economic price (0.55 $·mol–1), availability, and its dual function as

4

cosolvent and cosubstrate.15 However, this bienzymatic approach depends on both ADH and

5

aldehyde dehydrogenase, and acetic acid as the coproduct can cause a drop in reaction pH, which

6

requires additional pH control. Isopropanol is another commonly used cosubstrate with high

7

efficiency in the regeneration of NADH or NADPH.8 However, isopropanol must be applied in

8

significant molar excess, and the acetone produced needs to be removed to drive the reaction

9

forward and prevent enzyme inactivation.16 Consequently, there remains a constant demand for

10 11

new and improved cofactor regeneration approaches. A simple approach has been developed by Hollmann et al. by employing lactonizable diols

12

as “smart cosubstrates”.17 The preliminary step involves 1,4-butanediol (1,4-BD) reduced into 4-

13

hydroxybutanal (step 1), and then spontaneously converted into 2-hydroxytetrahydrofuran (2-

14

HTHF) (Scheme 1). The second step is an irreversible reduction into V #!

15

(step 2), with a T6D of –8.2 kcal·mol–1, which liberates two equivalents of NAD(P)H. In

16

addition, the final product, GBL, can be considered an important bulk chemical and is

17

thermodynamically stable and kinetically inert. This quality could shift the equilibrium and

18

thereby reduce the molar excess requirement of the cosubstrate.18 This novel regeneration

19

approach has been applied to NADH regeneration by employing HLADH, TeSADH, or evo

20

1.1.200.18–24 However, NADPH regeneration approaches have so far been insufficiently

21

developed, and only a few ADHs capable of regenerating NADPH using diols have been

22

reported. ADH from Lactobacillus brevis (LbADH) has been shown to be able to catalyze the

23

oxidation of 1,4-BD and 1,6-hexanediol (1,6-HD), but the low activity of the enzyme and


(GBL)

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

inhibition at 197 mM diols makes this system less than ideal.21 Considering many industrial

2

enzymes are NADPH-dependent, engineering ADHs for regenerating NADPH by oxidation of

3

diols is of particular interest.

4

An NADPH-dependent alcohol dehydrogenase (KpADH) was identified from

5

Kluyveromyces polysporus by genome mining, and was demonstrated to be capable of reducing

6

bulky-bulky ketones with moderate enantioselectivity.25 KpADH has been shown to follow the

7

ordered bi-bi kinetic mechanism and ‘Prelog’ priority in the reduction of prochiral ketones.26

8

Employing hydroclassified combinatorial saturation mutagenesis (HCSM) and polarity scanning

9

strategies, the e.e. values of KpADH were increased to 99.4% (R) and reversed to 97.8% (S) in

10

the reduction of 500 mM (4-chlorophenyl)(pyridine-2-yl)ketone.27,28 Unlike homologous

11

enzymes, such as Gre2p (51.3%) and CgKR1 (50.4%), KpADH exhibits high oxidative activity

12

toward secondary alcohols such as isopropanol, but has relatively lower activity toward 1,4-BD.

13

Moreover, the analysis of kinetic constants revealed that the oxidation of 1,4-BD is the rate-

14

limiting step, and there is a remarkable difference in kinetics between step 1 and step 2. Our

15

strategy for the regeneration of NADPH with diols as ‘smart cosubstrates’ was then focused on

16

improving the efficiency of KpADH toward 1,4-BD and balancing the kinetics between the two

17

steps. The feasibility of this newly developed NADPH regeneration enzyme was evaluated in the

18

preparation of D-amino acids.

19

EXPERIMENTAL SECTION

20

General remarks. 1,4-Butanediol (1,4-BD), 1,5-pentanediol (1,5-PD) and 1,6-

21

hexanediol (1,6-HD) were purchased from Aladdin (Shanghai) Co., Ltd. NADP+ and

22

NADPH were bought from Bontac Bio-engineering (Shenzhen) Co., Ltd. PPA was



1

purchased from Bodi Biotechnology (Shanghai) Co., Ltd. D-Phe was purchased from

2

CIVI Chemical Technology (Shanghai) Co., Ltd. Other reagents were purchased from

3

Sinopharm (Shanghai) Co., Ltd unless otherwise stated.

4

Construction of mutant library and high throughput screening. Saturation

5

mutagenesis library was constructed employing whole plasmid PCR with recombinant

6

pET28-kpadh coding for the alcohol dehydrogenase KpADH as a template. PCR

7

amplification system was as follows: 1.8 W* dNTP mixture, 0.2 W* PrimeSTAR® HS

8

DNA polymerase, 4 W* 5×PrimeSTAR® HS DNA polymerase buffer, 0.4 W* forward

9

primer, 0.4 W* reverse primer (Table S1), 0.4 W* pET28-kpadh and 12.4 W* ddH2O. PCR

10

program was: pre-denaturation at 94°C for 4 min, followed by amplification for 20 cycles

11

with denaturation at 98°C for 10 s, annealing at 55°C for 15 s and elongation at 72°C for

12

6 min, and then further elongation at 72°C for 10 min. The resultant PCR product was

13

digested with DpnI at 37°C for 0.5 h to eliminate the parental plasmid. Then, 10 W* of the

14

digestion mixture was transformed into Escherichia coli BL(DE3). Single colonies were

15

picked up and inoculated into 300 W* of LB medium (50 W ·mL–1 Kan) in each well of a

16

96-deep well plate and further incubated at 37°C for 12 h. Furthermore, 50 W* culture

17

was transferred into another 96-deep well plate containing 600 W* of LB in each well (50

18

W ·mL–1 of Kan). After incubation at 37°C and 120 rpm for 2.5 h, 0.2 mM IPTG was

19

added and further cultivated at 25°C and 120 rpm for 5 h. Cells were harvested by

20

centrifugation at 4000 rpm for 10 min and stored at –80°C. Then cells thawed and were

21

treated with 750 mg·L–1 of lysozyme at 37°C for 1 h. The cell lysate was collected by

22

centrifugation at 4000 rpm for 10 min.


Page 6 of 36

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1


Activity assay. The oxidative activities were determined by monitoring the increased

2

absorbance of NADPH at 340 nm with molar extinction coefficient of 6220

3

L·mol[ ·cm[ . The reaction mixture consisted of 10 mM diols, 1 mM NADP+ in Glycine-

4

NaOH buffer (pH 9.5, 100 mM) and 10 W* supernatant at 30°C for 3 min. One unit of

5

oxidative activity was defined as the amount of enzyme required to catalyze the produce

6

of 1 W8

NADPH per minute under the above-mentioned activity assay condition.

7

Rescreening of beneficial mutants. KpADH variants with higher activity were

8

inoculated into 40 mL LB medium, cultivated at 37°C and 120 rpm. Until OD600 reached

9

0.6~0.8, 0.2 mM IPTG was added and further cultivated at 25°C and 120 rpm for 6 h. The

10

cells were harvested by centrifugation at 8000 rpm and 4°C for 10 min, re-suspended with

11

phosphate buffer (100 mM, pH 7.0), and disrupted with nano-homogenizer (AH-BASIC-

12

I, ATS). The activities of variants toward different diols were measured as above

13

mentioned. Protein concentration was determined by the Bradford method with BSA as

14

standard protein.

15

Expression and protein purification of KpADH variants. To purify the WTKpADH

16

and variants, the collected cells were re-suspended in 10 mL of buffer A (20 mM

17

imidazole and 500 mM NaCl in 20 mM PBS buffer), and disrupted by ultra-sonication

18

(work for 1 s, pause for 3 s, 450 W). Cell debris were removed by centrifugation at 8000

19

rpm and 4°C for 30 min. Then, the supernatant was passed through 0.22 W8 filter and

20

loaded on the HisTrap HP nickel affinity column pre-equilibrated with buffer A using

21

AKTA Avant System (GE Healthcare Co., Ltd). Target proteins were radiantly eluted

22

employing buffer A and buffer B (500 mM imidazole and 500 mM NaCl in 20 mM PBS

23

buffer). Target proteins were collected and verified by SDS-PAGE analysis. WTKpADH



1

and variants were concentrated and mixed with 20

2

further use (Figure S3-S12).

3

Page 8 of 36

glycerol, and stored at –80°C for

Determination of kinetic parameters. Kinetic parameters of the purified WTKpADH

4

and variants were determined employing the standard activity assay method. Different

5

concentrations of 1,4-BD (20.0–250 mM) or 2-HTHF (1.0–25.0 mM), NADP+ (0.01–0.60

6

mM) and appropriate of purified enzymes were mixed in pH 9.5 Glycine-NaOH buffer

7

and 30°C. All the activities were determined in triplicate. The KM, Vmax, kcat and Kia were

8

calculated according to the simulation based on Michaelis-Menten double substrate

9

kinetics employing Matlab software package (Figure S17-S18).

10

Application of KpADHV84I/Y127M in NADPH regeneration with 1,4-BD as a ‘smart

11

cosubstrate’. Reaction mixtures for the comparison of different NADPH regeneration

12

systems consisted of 50 mM 1,4-BD or 100 mM glucose, 0.5 mM NADP+, 50 mM PPA,

13

100 mM ammonium sulfate, 1.5 kU·L–1 purified WTKpADH, KpADHV84I, KpADHY127C,

14

KpADHV84I/Y127C and KpADHV84I/Y127M or glucose dehydrogenase in 200 mM Gly-NaOH

15

(pH 9.0). Reactions were started by the addition of DAADHD94A partially purified by

16

heating and incubated at 30°C and 120 rpm. Aliquot samples (100 W*) were withdrawn at

17

intervals and the reaction was stopped by heating at 100°C for 10 min, followed by

18

centrifugation at 12000 rpm for 5 min. The formation of D-Phe and V #!

19

monitored by HPLC and GC.

20

was

Effect of 1,4-BD addition and PPA concentrations on the reaction were conducted with

21

50 mM or 100 mM PPA, 25 mM or 50 mM 1,4-BD, 100 mM or 200 mM ammonium

22

sulfate, 0.5 mM NADP+, 1.5 kU·L–1 purified KpADHV84I/Y127M in 200 mM Gly-NaOH


Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

(pH 9.0). The reactions were also started and monitored as above described. TOF was

2

defined as one W8

3

KpADHV84I/Y127M and determined over the first half hour.

4

of D-Phe formed in one minute by using one W8

of

Analytical Methods. D-Phe was determined by high-performance liquid

5

chromatography (HPLC) using Astec CHIROBIOTICTM T column (150 mm × 4.6 mm ×

6

5 W8 Sigma Technologies Co. Ltd) with methanol and ultrapure water (70:30, v/v) as

7

mobile phase. The retention time of D-Phe was 8.49 min at 210 nm and 30°C with a flow

8

rate of 0.5 mL·min–1 (Figure S15). GBL was measured by gas chromatography (GC)

9

equipped with CP-Chirasil-Dex CB (25 m × 0.25 mm × 0.25 W8 Agilent Technologies

10

Co. Ltd ). The column temperature was initially kept at 70°C for 5 min, and increased to

11

200°C at 20°C·min–1 and kept for 5 min. The injector and detector temperatures were

12

250°C. The retention time of GBL was 7.92 min (Figure S16).

13

Molecular docking and molecular dynamic analysis. Docking runs were carried

14

out using standard parameters of the program for interactive growing and subsequent

15

scoring. Molecular dynamic simulation was carried out employing the NAMD module of

16

Discover Studio 3.5. CHARMm forcefield and temperature of 300 K were adopted, and

17

the simulation time was set as 20 ns. Binding energies were calculated employing the

18

internal scoring module of Discovery Studio 3.5.27,28

19

Determination of optimum pH and stability of WTKpADH and KpADHV84I/Y127M.

20

Effect of pH on the activities of purified WTKpADH and KpADHV84I/Y127M were

21

determined employing above mentioned standard activity assay except for different

22

buffers with pH ranging from 6.0 to 8.0 (PBS buffer) and 8.0 to 10.0 (Glycine-NaOH



1

buffer). Purified WTKpADH and KpADHV84I/Y127M (1.0 mg·mL–1) were incubated in

2

buffers of pH 9.0 and 9.5 at 30°C. Aliquot samples (10 W*G were withdrawn at intervals,

3

and activities were determined employing the standard activity assay method. All

4

activities were determined in triplicate.

5

RESULTS AND DISCUSSION

6

Identifying key residues of KpADH by MD simulations.

7

The kinetic parameters of purified KpADH toward 1,4-BD and 2-HTHF were determined in

8

order to better understand its enzymatic characteristics. The KM, kcat, and kcat/KM of KpADH

9

toward 1,4-BD were 50.6 mM, 8.76 min–1, and 0.173 min–1·mM–1 (Table 2), respectively, while

10

the KM, kcat, and kcat/KM toward 2-HTHF were 2.23 mM, 2.63×103 min–1, and 1.18×103 min–

11

1·mM–1 (Table

12

kcat/KM of this step was undesirable for the efficient regeneration of NADPH. Remarkably, the

13

kcat/KM toward 2-HTHF was 6.82×103-fold higher than that of 1,4-BD, and the unmatched

14

kcat/KM values between step 1 and step 2 were also unfavorable for the NADPH regeneration.

15

Consequently, KpADH was engineered to increase its kcat/KM value toward 1,4-BD, and to

16

balance the kinetics between 1,4-BD and 2-HTHF.

4), respectively. The oxidation of 1,4-BD was the rate-limiting step, and the low

17

KpADH displays low sequence similarity to the enzymes deposited in NCBI, with the

18

highest identity of only 54.5%. Moreover, homologous enzymes, such as Gre2p and CgKR1,

19

could not catalyze the oxidation of isopropanol or 1,4-BD.29,30 Fortunately, the crystal structure

20

of wild type KpADH docked with 1,4-BD was available (WTKpADH, PDB: 5Z2X), and this

21

structure was used in a molecular dynamic (MD) simulation of 20 ns to identify key residues that

22

may affect the binding and catalytic activity of KpADH toward 1,4-BD. The average


Page 10 of 36

Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

conformation was extracted as shown in Figure 1. Distance analysis demonstrated that 1,4-BD

2

was surrounded by residues V84, Y127, S196, F197, V198, E214, and T215. The distances

3

between 1,4-BD and residues V84, Y127, and S196 were less than 3.0 Å. Notably, residue Y127,

4

which is 2.2 Å from 1,4-BD, might produce significant steric hindrance. Consensus analysis of

5

the above mentioned residues was carried out for ADHs of various origins (Figure S1). With the

6

exception of a strictly conserved residue T215, the other 6 residues were analyzed for their roles

7

in binding and oxidizing of 1,4-BD.

8

Engineering KpADH for improved activity toward 1,4-BD

9

Each residue was subjected to saturation mutagenesis and high throughput screening with

10

1,4-BD as the substrate and NADP+ as a cofactor (Figure S2). All variants of V198 and E214

11

displayed decreased activity toward 1,4-BD compared with WTKpADH. The variants that showed

12

the highest potential activity in each library were KpADHV84I, KpADHY127C, KpADHY127M,

13

KpADHS196A, and KpADHF197W, with relative activities of 392%, 261%, 248%, 118%, and

14

120%, respectively (Figure 2). The relative activities of purified KpADHV84I, KpADHY127C,

15

KpADHY127M, KpADHS196A, and KpADHF197W toward isopropanol were 21.7%, 22.8%, 105%,

16

93.0%, and 19.6%, respectively. Only KpADHY127M exhibited increased activities toward both

17

1,4-BD and isopropanol. Variant KpADHV84I exhibited the highest activity for the oxidation of

18

1,4-BD among all single mutants.

19

In order to gain insight into the binding affinity and catalytic efficiency of the variants, their

20

kinetic parameters were determined according to the ordered bi-bi kinetic model.31 As shown in

21

Table 2, KpADHV84I displayed a kcat of 80.3 min–1, which was 9.17-fold higher than WTKpADH,

22

suggesting residue 84 is vital for the oxidation activity toward 1,4-BD. However, a higher KM



1

toward 1,4-BD and NADP+ was observed in KpADHV84I, 1.71- and 1.96-fold higher than that of

2

WTKpADH,

3

BD and NADP+. As a result, the kcat/KM of KpADHV84I toward 1,4-BD was 5.36-fold higher than

4

WTKpADH.

5

respectively, which is suggestive of a weaker substrate binding affinity toward 1,4-

The kcat values of KpADHY127C and KpADHY127M were 2.07- and 3.70-fold higher than that

6

of WTKpADH, and their apparent KM values toward 1,4-BD were 21.4 mM and 28.9 mM,

7

respectively, both of which were lower than that of WTKpADH (50.6 mM), KpADHY127C and

8

KpADHY127M also displayed a decrease KM toward NADP+ (52.8% and 93.0%) in contrast with

9

that of WTKpADH. The kinetics data demonstrated that residue Y127 was critical in both

10

catalytic efficiency and binding of 1,4-BD and NADP+. The kcat/KM of KpADHY127M toward 1,4-

11

BD was 1.12 min–1·M–1, a 6.47-fold increase over that of WTKpADH. The kcat/KM of

12

KpADHS196A and KpADHF197W were improved to 0.301 min–1·mM–1 and 0.673 min–1·mM–1,

13

respectively. Based on the above results, residue V84 appears to play an important role in

14

catalytic efficiency, while Y127 seems crucial for both catalytic efficiency and binding of 1,4-

15

BD and NADP+ in the rate-limiting step.

16

Combinatorial mutagenesis was then performed to further improve the activity of the

17

selected variants KpADHV84I/Y127C, KpADHV84I/Y127M, KpADHY127C/S196A, and KpADHS196A/F197W

18

(Figure 2). The relative activities of KpADHV84I/Y127C and KpADHV84I/Y127M toward 1,4-BD were

19

358% and 271%, while toward isopropanol they were 43.7% and 126%, respectively. Variant

20

KpADHV84I/Y127C exhibited the highest activity among all double mutants, but still had lower

21

activity than that of KpADHV84I. It is worth mentioning that KpADHV84I/Y127M displayed

22

improved activity toward both 1,4-BD and isopropanol, similar to KpADHY127M. Considering the

23

limited effect of the F197W mutation on oxidative activity and the negative effect of the S196A


Page 12 of 36

Page 13 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

mutation on thermal stability (data not shown), triple combinatorial mutagenesis was not

2

explored.

3

Due to its decreased KM (31.0 mM) and increased kcat (28.7 min–1) toward 1,4-BD, the

4

kcat/KM of the double mutant KpADHY127C/S196A was 0.926 min–1·mM–1 (Table 2), which was

5

1.09- and 3.08-fold above those of KpADHY127C and KpADHS196A, respectively, indicating a

6

synergistic effect of the Y127C and S196A mutations. With regard to variant KpADHS196A/F197W,

7

a decreased KM (23.4 mM) and a slightly enhanced kcat (10.2 min–1) were found, giving a kcat/KM

8

toward 1,4-BD of 0.436 min–1·mM–1, the combination of two adjacent residues (S196A and

9

F197W) did not play a cooperative role. Compared with KpADHV84I and KpADHY127C, double

10

mutant KpADHV84I/Y127C displayed a further increased kcat/KM of 1.30 min–1·mM–1, which was

11

7.51-fold above that of WTKpADH. As shown in Table 2, the highest kcat/KM value toward 1,4-

12

BD of 2.00 min–1·mM–1 was observed with KpADHV84I/Y127M, 11.6-fold higher than that of

13

WTKpADH.

14

kcat value (8.76 min–1 to 30.1 min–1) contributed to the remarkably improved kcat/KM of

15

KpADHV84I/Y127M toward 1,4-BD. Owing to the positive effect of Y127M mutation on substrate

16

binding, the KM of KpADHV84I/Y127M toward 1,4-BD was over 5 times lower than that of

17

KpADHV84I. In addition, KpADHV84I/Y127M also showed the best affinity toward NADP+, the KM

18

toward NADP+ decreased from 15.8 mM (WTKpADH) to 3.05 mM. Consequently, Y127 was

19

shown to be a key residue that could coordinate with other residues in enhancing the binding

20

affinity of substrate. Importantly, no substrate inhibition was detected in all tested single and

21

double mutants, even at 250 mM 1,4-BD. The Kia values, dissociation constants toward NADP+,

22

of most mutants improved to varying degrees, indicating that increased binding to non-catalytic

Both a significant decrease in KM value (50.6 mM to 15.1 mM) and an increase in



1

site, but there was no significant effect on affinity of NADP+ observed from the apparent KM

2

toward NADP+, in the rate-limiting step for oxidation of 1,4-BD.

3

Analysis of the substrate specificities of WTKpADH and variants

4

Kinetic parameters of WTKpADH and other variants toward 1,5-pentanediol (1,5-PD) and

5

1,6-hexanediol (1,6-HD) were also investigated (Table 3). The kcat/KM of WTKpADH toward 1,5-

6

PD and 1,6-HD were 0.75 min–1·mM–1 and 2.27 min–1·mM–1, respectively, representing 4.34-

7

and 13.1-fold increases compared 1,4-BD. The result suggests that diols with longer chain length

8

than 1,4-BD could be more efficiently oxidized by KpADH. Compared with 1,4-BD, the KM

9

values of KpADHV84I toward 1,5-PD and 1,6-HD were reduced by 50.9% (42.5 mM) and 85.8%

10

(12.3 mM), respectively, indicating the positive role of KpADHV84I in enhancing the binding

11

affinity of longer chain diols. However, KpADHY127C had a greatly decreased kcat/KM and was

12

found to be unfavorable for the oxidation of 1,5-PD and 1,6-HD. Unlike KpADHY127C, the

13

KpADHY127M displayed a remarkable effect on binding affinity. The KM values of KpADHY127M

14

toward 1,5-PD and 1,6-HD were decreased to 14.3 mM and 6.50 mM, respectively, which

15

represented a reduction of 80.8% and 64.9% compared to those of WTKpADH. As a result, the

16

kcat/KM of KpADHY127M toward 1,5-PD and 1,6-HD reached 3.27 min–1·mM–1 and 6.09 min–

17

1·mM–1,

18

respectively, which were 4.36- and 2.68-fold higher than WTKpADH.

Mutations of S196 into Ala and F197 into Trp had little influence on the oxidization of 1,5-

19

PD and 1,6-HD. With regard to double mutant KpADHV84I/Y127M, the KM values toward 1,5-PD

20

and 1,6-HD were further decreased to 8.20 mM and 6.50 mM, respectively, which were merely

21

11.0% and 35.1% of those of WTKpADH. The corresponding kcat/KM values were 2.35 min–

22

1·mM–1

and 11.5 min–1·mM–1. These values represented 3.13- and 5.07-fold increases over


Page 14 of 36

Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

WTKpADH.

2

observed sequence of 1,6-HD>1,5-PD>1,4-BD, which was consistent with WTKpADH. Only

3

CDX-019, LbADH, and TeSADH have been reported to be able to catalyze the oxidation of 1,6-

4

HD for NAPDH regeneration, with Vmax of 0.1, 0.4, and 0.4 W8 M8

5

The Vmax values of KpADHV84I/Y127C and KpADHV84I/Y127M were calculated to be 2.34 and 1.79

6

W8 M8

7

remarkable kinetic parameters, variant KpADHV84I/Y127M could be a promising NADPH

8

regeneration system utilizing diols as ‘smart cosubstrates’.

9

Comparison of different cofactor regeneration systems

Interestingly, the kcat/KM of KpADHV84I/Y127M toward tested diols obeyed the

–1·mg–1,

–1·mg–1,

respectively.21

which were almost 5 times higher than the highest reported. Based on its

10

Cofactor regeneration systems with diols as ‘smart cosubstrates’ have been successfully

11

demonstrated in the coupled reactions catalyzed by NADH-dependent enoate reductase from

12

Thermus scotoductus SA-01 (TsER), 3-hydroxybenzoate-6-hydroxylase from Rhodococcus jostii

13

RHA1 (3HB6H), and Baeyer–Villiger monooxygenase from Acinetobacter sp. NCIMB 9871

14

(BVMO).18, 21 Herein, the application potential of this newly developed NADPH regeneration

15

system was evaluated by coupling it to a NADPH-dependent amino acid dehydrogenase from

16

Ureibacillus thermosphaericus (DAADHD94A).32 DAADHD94A catalyzed the asymmetric

17

reduction of phenylpyruvic acid (PPA) into D-phenylalanine (D-Phe), an important chiral

18

building block for nateglinide (a drug for type II diabetes).33

19

Cofactor regeneration with 1,4-BD and glucose as cosubstrates were initially tested at 50

20

mM PPA. 1,4-BD and glucose were added in 1.0-equivalent and 2.0-equivalents relative to the

21

substrate. WTKpADH, KpADHV84I/Y127M, and glucose dehydrogenase (GDH) were added to the

22

level of equivalent activities (1.5 kU·L–1). As illustrated in Figure 3A, the yield of D-Phe



1

reached 72.1±1.9% after 1 h and slightly increased to 74.4±1.6% after 4 h employing GDH for

2

cofactor regeneration. Without pH adjustment, the pH of the reaction declined to lower than 6.0

3

because of the gluconic acid produced, which was disadvantageous for the reaction since the

4

optimal pH of DAADHD94A in the asymmetric reduction was pH 9.0. The conversion curve

5

employing KpADHV84I/Y127M was distinct from that of WTKpADH. With WTKpADH, the yield of

6

D-Phe

7

reaction time. For KpADHV84I/Y127M, the initial reaction rate was so high that the conversion

8

reached 98.5±2.3% in 1 h. PPA was completely converted into D-Phe in 4 h. As previously

9

mentioned, the conversion of 1,4-BD into GBL is irreversible and thermodynamically favored.

10

Moreover, the optimum pH for both WTKpADH and KpADHV84I/Y127M was determined to be pH

11

9.5 (Figure S19), which was much more favorable for DAADHD94A than GDH. Also,

12

KpADHV84I/Y127M exhibited greater stability than WTKpADH at pH 9.5 (Figure S20).

13

was 46.4±1.9% at 4 h, and a slightly higher conversion of 50% was obtained after further

The kcat/KM of KpADHV84I/Y127M toward 2-HTHF in step 2 was 195 min–1·mM–1

14

(Table 4). It is worth noting that the ratio of kcat/KM toward 2-HTHF and 1,4-BD was

15

97.8, which is nearly 70-fold lower than that of WTKpADH (kcat/KM of 6.82×103). To

16

investigate the significance of the ratio of kcat/KM between 2-HTHF and 1,4-BD, variants

17

KpADHV84I/Y127C, KpADHY127C, and KpADHV84I (kcat/KM ratios of 51.6, 228, and 775)

18

were also applied as NADPH regeneration systems at the same enzyme loading of 1.5

19

kU·L–1. A higher yield of 83.7% was observed with KpADHV84I/Y127C at 1 h, which might

20

be due to its higher kcat/Km toward 1,4-BD. Similar yields were obtained at 1 h (61.6%

21

and 67.6%) with KpADHY127C and KpADHV84I, which exhibited similar kcat/KM toward

22

1,4-BD. Among variants with kcat/KM ratios of 51.6 6.82×103 tested, KpADHV84I/Y127M

23

possessed a kcat/KM ratio of around 100 and was shown to be an appropriate reaction


Page 16 of 36

Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

system for the sequential oxidation of 1,4-BD into GBL. Consequently, both improved

2

catalytic efficiency and balanced kcat/KM between 1,4-BD and 2-HTHF contributed to the

3

high efficiency in reductive amination of PPA into D-Phe.

4

Application of KpADHV84I/Y127M in the synthesis of D-Phe

5

Since the sequential oxidation of one 1,4-BD molecule produced two equivalents of

6

NADPH, the addition of 1,4-BD was further reduced to 0.5-equivalents of the substrate in the

7

KpADHV84I/Y127M NADPH regeneration system. As shown in Figure 3B, the initial production of

8

D-Phe

9

mM) (Figure 3A), which indicated that a higher amount of cosubstrate was advantageous for

was 67.6±1.5% at 0.5 h, 20.7% lower than that with 1,4-BD at 1.0-equivalent of PPA (50

10

NADPH regeneration. With 25 mM 1,4-BD, the yield of D-Phe reached 99.8±1.0% in 2 h at 50

11

mM PPA. The production of the coproduct GBL, an important solvent, was determined to be

12

22.8 mM at 4 h. As a result, in comparison to the substrate, 0.5-equivalents of 1,4-BD was

13

adequate for the bioreduction system. When PPA loading was further increased to 100 mM, the

14

addition of 1,4-BD was also increased to 50 mM. The yield of D-Phe increased rapidly to

15

95.0±1.7% in the first 3 h. The production of D-Phe then slowly increased to 99.2±1.5% after 6

16

h, along with 45.2 mM GBL produced. Enantioselectivity analysis revealed that the e.e. value of

17

D-Phe

18

calculated to be 3.7 minb , which was a TOF value 2.0-fold higher than TeSADH using 1,6-

19

hexanediol as a cosubstrate at the same concentration.21 Considering the basic pH preference of

20

DAADHD94A, the traditional strategy was found to be at the expense of a high loading of GDH.

21

The highest observed conversion for preparing D-Phe was achieved with a variant of meso-

22

DAPDH from Symbiobacterium thermophilum. As much as 100 mM PPA was converted with an

23

efficiency of 96.9% conversion after 24 h with the assistance of 0.2 g·L–1 GDH (c.a. 36 KU·L–

produced was over 99.9%. The turnover frequency (TOF) value of KpADHV84I/Y127M was



1

1).34,35

2

more advantageous in NADPH regeneration for amino acid dehydrogenase. Optimizing on the

3

regeneration of NADPH with 1,6-HD is also undergoing, since KpADHV84I/Y127M display much

4

higher efficiency than 1,4-BD.

5

Elucidating the mechanism of enhanced activity toward 1,4-BD

6

This newly developed cofactor regeneration system employing KpADHV84I/Y127M was

Virtual mutation and MD simulation were conducted to explore the molecular basis of the

7

improved activity and balanced kcat/KM. After MD simulation of 20 ns, the average

8

conformations were extracted. The interactions and binding energies were also calculated. In the

9

conformation of WTKpADH in complex with 1,4-BD, only one hydrogen bond between Tyr164

10

and 1,4-BD was identified (Figure 4A). It is presumed that the bulky Tyr127 residue might

11

block the binding and rotation of 1,4-BD. Moreover, no interaction was found between Ser126

12

and 1,4-BD, and Ser126 had been shown to be essential for stabilizing the substrate.36,37

13

However, more hydrogen bonds were predicted in the KpADHV84I/Y127M complex, although no

14

directed interaction was found between 1,4-BD and KpADHV84I/Y127M. The substrate binding

15

pocket of KpADHV84I/Y127M had a tendency to become a little larger and more hydrophobic than

16

that of WTKpADH. The enhanced hydrophobicity of Ile84 was shown to have pushed 1,4-BD

17

toward the bottom of substrate binding pocket, while the mutation of Y127M increased the

18

binding pocket volume. These changes gave rise to the two possible extra hydrogen bonds

19

between Ser126 (or Ala128) and the hydroxyl group of 1,4-BD (Figure 4B, detailed interactions

20

could be found in Figure S13 and Figure S14). The binding energy between KpADHV84I/Y127M

21

and 1,4-BD was calculated to be –81.6 kcal·mol–1, much lower than –70.0 kcal·mol–1 of

22

WTKpADH.

23

hydrogen bonds with Ser126 or Tyr164, .-alkyl interaction with Phe197, and alky-alkyl

Interactions between KpADH and 2-HTHF could be found in Figure 4C, including


Page 18 of 36

Page 19 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

interaction with Ala128. The intricate binding network between KpADH and 2-HTHF led to a

2

high kcat/KM toward 2-HTHF. The binding energy was calculated to be –102.7 kcal·mol–1. The

3

altered hydrophobicity of the substrate binding pocket also resulted in the movement of 2-HTHF

4

toward the bottom. Due to the longer length of 2-HTHF than 1,4-BD, the movement led to an

5

extended distance (4.1 Å) between Ser126 and 2-HTHF, which was unfavorable for the

6

formation of hydrogen bonds. However, other interactions between Tyr164, Ala128, and Phe197

7

were still present (Figure 4D). As a result, the binding energy between variant KpADHV84I/Y127M

8

and 2-HTHF was reduced to –89.3 kcal·mol–1, and the kcat/KM was decreased, as shown in Table

9

4. As mentioned above, the mutation of Val84 and Tyr127 into Ile and Met altered the spatial

10

structure and hydrophobicity of the substrate binding pocket. The substrates might better access

11

the bottom of the substrate binding pocket with changes in hydrogen-bonding network, varying

12

from 1,4-BD and 2-HTHF.

13

CONCLUSIONS

14

The alcohol dehydrogenase KpADH was engineered for improved NADPH

15

regeneration employing 1,4-BD as a ‘smart cosubstrate’. Key residues were identified by

16

molecular dynamics and subjected to saturational and combinatorial mutagenesis.

17

Residues V84 and Y127 were shown to play essential roles in the catalytic efficiency and

18

substrate binding affinity, and mutations to these key residues favorably altered the

19

substrate binding pocket and enhanced the hydrogen bonding network between the

20

mutant enzyme and 1,4-BD. In the rate-limiting step, the kcat/KM of KpADHV84I/Y127M

21

toward 1,4-BD was 2.00 min–1·mM–1, 11.6-fold higher than WTKpADH. Moreover, the

22

ratio of kcat/KM toward 2-HTHF versus 1,4-BD was dramatically reduced to 97.8 from



1

6.82×103 in WTKpADH, which was favorable for the sequential oxidation of 1,4-BD.

2

Variant KpADHV84I/Y127M was applied in a coupled reaction for NADPH regeneration. As

3

much as 100 mM PPA was converted into D-Phe with a 99.2% yield employing 50 mM

4

1,4-BD. This enhancement was ascribed to the improved catalytic efficiency toward 1,4-

5

BD and the balanced kcat/KM between 1,4-BD and 2-HTHF. This study has demonstrated

6

the engineering and utilization capacity of ADH as an efficient NADPH regeneration

7

system employing diols as ‘smart cosubstrates’.

8

ASSOCIATED CONTENT

9

Supporting Information.

10

The Supporting Information is available free of charge on the ACS Publications website.

11

Table S1, primers used in this study; Figure S1, consensus analysis; Figure S2, HTS results;

12

Figure S3bS12, purification of variants; Figure S13bS14, substrate binding pocket analysis;

13

Figure S15bS16, calibration curves; Figure S17bS18, kinetic parameters analysis by Matlab

14

software; Figure S19b-

15

KpADHV84I/Y127M.

16

AUTHOR INFORMATION

17

Corresponding Author

18

* Email for Y. Ni: [email protected]

19

Author Contributions

20

¶(G.C.

21

Notes

Optimal pH and thermal stability of WTKpADH and

Xu, C. Zhu) These authors contributed equally to this work.


Page 20 of 36

Page 21 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

The authors declare no competing financial interest.

2

ACKNOWLEDGMENT

3

We are grateful to the National Key Research and Development Program (2018YFA0901700),

4

the National Natural Science Foundation of China (21506073, 21776112), the national first-class

5

discipline program of Light Industry Technology and Engineering (LITE2018-07), the Natural

6

Science Foundation of Jiangsu Province (BK20171135), the Program of Introducing Talents of

7

Discipline to Universities (111-2-06), and Top-notch Academic Programs Project of Jiangsu

8

Higher Education Institutions for the financial support of this research.

9

ABBREVIATIONS

10

1,4-BD, 1,4-butanediol; 2-HTHF, 2-hydroxytetrahydrofuran; GBL, V #!

11

pentanediol; 1,6-HD, 1,6-hexanediol; TOF: turnover frequency; PPA, phenylpyruvic acid; D-

12

Phe: D-phenylalanine.

13

REFERENCES

14

(1) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.; Robins, K.

15

Engineering the Third Wave of Bocatalysis. Nature 2012, 485, 9)b : 4

16

(2) Sheldon, R. A.; Woodley, J. M. Role of Biocatalysis in Sustainable Chemistry. Chem. Rev.

17

2018, 118, 9; b9 94

18

(3) Savile, C. K.; Janey, J. M.; Mundorff, E. C.; Moore, J. C.; Tam, S.; Jarvis, W. R.; Colbeck, J.

19

C.; Krebber, A.; Fleitz, F. j.; Brands, J.; Devine, P. N.; Huisman, G. W.; Hughes, G. J.

20

Biocatalytic Asymmetric Synthesis of Chiral Amines from Ketones Applied to Sitagliptin

21

Manufacture. Science 2010, 329, ;)b ;:4


A 1,5-PD, 1,5-


1

(4) Hollmann, F.; Arends, I. W. C. E.; Holtmann, D. Enzymatic Reductions for the Chemist.

2

Green Chem. 2011, 13,

3

(5) Cheng, J.; Xu, G. C.; Han, R. Z.; Dong, J. J.; Ni, Y. Efficient Access to L-Phenylglycine

4

Using a Newly Identified Amino Acid Dehydrogenase from Bacillus clausii. RSC Adv. 2016, 6,

5

9;))5b9;). 4

6

(6) Hollmann, F.; Arends, I. W. C. E.; Buehler, K.; Schallmey, A.; Buehler, B. Enzyme-

7

Mediated Oxidations for the Chemist. Green Chem. 2011, 13,

8

(7) Magnao, J., Dunetz, J. R., Large-scale Carbonyl Reductions in the Pharmaceutical Industry.

9

Org. Process Res. Dev. 2012, 16,

9)b

4

).b

.b .)4

9 4

10

(8) Huisman, G. W.; Liang, J.; Krebber, A. Practical Chiral Alcohol Manufacture Using

11

Ketoreductases. Curr. Opin. Chem. Biol. 2010, 14,

12

(9) Zhao, H. M.; van der Donk, W. A. Regeneration of Cofactors for Use in Biocatalysis. Curr.

13

Opin. Biotechnol. 2003, 14, )9 b)9:4

14

(10) Kara, S.; Schrittwieser, J. H.; Hollmann, F.; Ansorge-Schumacher, M. B. Recent Trends and

15

Novel Concepts in Cofactor-Dependent Biotransformations, Appl. Microbiol. Biotechnol. 2014,

16

98, ) 5b ) :4

17

(11) Moore, J. C.; Pollard, D. J.; Kosjek, B.; Devine, P. N. Advances in the Enzymatic Reduction

18

of Ketones. Acc. Chem. Res. 2007, 40,

19

(12) Liu, W.; Wang, P. Cofactor Regeneration for Sustainable Enzymatic Biosynthesis.

20

Biotechnol. Adv. 2007, 25, .:b 9 4

b

b

:4

:4


Page 22 of 36

Page 23 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

(13) Ni, Y.; Holtmann, D.; Hollmann, F. How Green is Biocatalysis? To Calculate is to Know,

2

ChemCatChem. 2014, 6, : ;b:

3

(14) Hummel, Werner; Harald Gröger. Strategies for regeneration of nicotinamide coenzymes

4

emphasizing self-sufficient closed-loop recycling systems. Journal of biotechnology. 2014, 191

5

b

4

4

6

(14) Aksu, S.; Arends, I. W. C. E.; Hollmann, F. A New Regeneration System for Oxidized

7

Nicotinamide Cofactors, Adv. Synth. Catal. 2009, 351,

8

(15) Broussy, S.; Cheloha, R. W.; Berkowitz, D. B. Enantioselective, Ketoreductase-Based Entry

9

into Pharmaceutical Building Blocks: Ethanol as Tunable Nicotinamide Reductant. Org. Lett.

b

.4

10

2009, 11, ;)b ;94

11

(16) Schroer, K.; Tacha, E.; Luetz, S. Process Intensification for Substrate-Coupled Whole cell

12

Ketone Reduction by in situ Acetone Removal. Org. Process Res. Dev. 2007, 11, 9 .b9

13

(17) Kara, S.; Spickermann, D.; Schrittwieser, J. Weckbecker, H. A.; Laggewie, C.; Arends, I.

14

W.; Hollmann, F. Access to Lactone Building Blocks via Horse Liver Alcohol Dehydrogenase-

15

Catalyzed Oxidative Lactonization, ACS Catal. 2013, 3,

16

(18) Kara, S.; Spickermann, D.; Schrittwieser, J. H.; Leggewie, C.; van Berkel, W. J. H.; Arends,

17

I. W. C. E.; Hollmann, F. More Efficient Redox Biocatalysis by Utilizing 1,4-Butanediol as a

18

‘Smart Cosubstrate’, Green Chem. 2013, 15,

19

(19) Wachtmeister, J.; Jakoblinnert, A.; Rother, D. Stereoselective Two-Step Biocatalysis in

20

Organic Solvent: Toward All Stereoisomers of a 1,2-Diol at High Product Concentrations, Org.

21

Process Res. Dev. 2016, 20, 5

;b

.b

4

:4

)4

b 5) 4



Page 24 of 36

1

(20) Kara, S.; Spickermann, D.; Weckbecker, A.; Leggewie, C.; Arends, I. W. C. E.; Hollmann,

2

F. Bioreductions Catalyzed by an Alcohol Dehydrogenase in Non-Aqueous Media,

3

ChemCatChem 2014, 6, :5 b:5.4

4

(21) Bornadel, A.; Hatti-Kaul, R.; Hollmann, F.; Kara, S. A Bi-enzymatic Convergent Cascade

5

for e

6

(22) Huang, L.; Romero, E.; Ressmann, A. K.; Rudroff, F.; Hollmann, F.; Fraaije, M. W.; Kara,

7

S. Nicotinamide Adenine Dinucleotide-Dependent Redox-Neutral Convergent Cascade for

8

Lactonizations with Type II Flavin-Containing Monooxygenase. Adv. Synth. Catal. 2017, 359,

9

.

b

3

as a ‘Double-Smart Cosubstrate’. ChemCatChem 2015, 7,

b

)4

94

10

(23) Huang, L.; Sayoga, G. V.; Hollmann, F.; Kara, S. Horse Liver Alcohol Dehydrogenase-

11

Catalyzed Oxidation Lactamization of Amino Alcohols. ACS Catal. 2018, 8, 9.9;b9.9 4

12

(24) Zuhse, R.; Leggewie, C.; Hollmann, F.; Kara, S. Scaling-up of “smart cosubstrate” 1,4-

13

butanediol promoted asymmetric reduction of ethyl-4,4,4-trifluoroacetoacetate in organic media.

14

Organic Process Research & Development. 2015, 19, .:b 5 4

15

(25) Zhou, J. Y.; Xu, G. C.; Han, R. Z.; Dong, J. J.; Zhang, W. G.;Zhang, R. Z.; Ni, Y. Carbonyl

16

Group-Dependent High-Throughput Screening and Enzymatic Characterization of Diaromatic

17

Ketone Reductase. Catal. Sci. Technol. 2016, 6, .

18

(26) Cahn, R. S.; Ingold, S. C.; Prelog, V. Specification of Molecular Chirality. Angew. Chem.

19

Int. Ed. 1966, 5, 9)b

20

(27) Xu, G. C.; Wang, Y.; Tang, M. H.; Zhou, J. Y.; Zhao, J.; Han, R. Z; Ni, Y. Hydroclassified

21

Combinatorial Saturation Mutagenesis: Reshaping Substrate Binding Pockets of KpADH for

22

Enantioselective Reduction of Bulky–Bulky Ketones. ACS Catal. 2018, 8, 9

;[.

54

)4


.b9

)4

Page 25 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

(28) Zhou, J. Y; Wang, Y.; Xu, G. C; Wu, L.; Han, R. Z.; Schwaneberg, U.; Ni, Y. Structural

2

Insight into Enantioselective Inversion of an Alcohol Dehydrogenase Reveals a “Polar Gate” in

3

Stereorecognition of Diaryl Ketones. J. Am. Chem. Soc. 2018, 140,

4

(29) Park, H. J.; Jung, J.; Choi, H.; Uhm, K. N.; Kim, H. K. Enantioselective Bioconversion

5

Using Escherichia coli Cells Expressing Saccharomyces cerevisiae Reductase and Bacillus

6

subtilis Glucose Dehydrogenase. J. Microbiol. Biotechnol. 2010, 20,

7

(30) Qin, F. Y.; Qin, B.; Mori, T.; Wang, Y.; Meng, L. X.; Zhang, X.; Jia, X.; Abe, I.; You, S.

8

Engineering of Candida glabrata Ketoreductase 1 for Asymmetric Reduction of alpha-Halo

9

Ketones. ACS Catal. 2016, 6, .

)b.

. )b

;;b

.) 4

;.4

;4

10

(31) Wratten, C. C.; Cleland, W. W. Product Inhibition Studies on Yeast and Liver Alcohol

11

Dehydrogenases. Biochemistry 1963, 2, : )b:

12

(32) Hayashi, J. J.; Seto, T.; Akita, H.; Watanabe, M.; Hoshino, T.; Yoneda, K.; Ohshima, T.;

13

Sakuraba, H. Structure-based Engineering of an Artificially Generated NADP+-Dependent D-

14

amino acid Dehydrogenase. Appl. Environ. Microbiol. 2017, 83, e00491-17.

15

(33) White, J. R.; Campbell, R. K. Recent Developments in the Pharmacological Reduction of

16

Blood Glucose in Patients with Type 2 Diabetes. Clin. Diabetes 2001, 19, ) b ):4

17

(34) Gao, X. Z.; Huang, F.; Feng, J. H.; Chen, X.; Zhang, H. L.; Wang, Z. X.; Wu, Q. Q.; Zhu,

18

D. M. Engineering the meso-Diaminopimelate Dehydrogenase from Symbiobacterium

19

thermophilum by Site Saturation Mutagenesis for D-Phenylalanine Synthesis. Appl. Environ.

20

Microbiol. 2013, 79, );5:b);9 4

4



1

(35) Makino, Y.; Negoro, S.; Urabe, I.; Okada, H. Stability-Increasing Mutants of Glucose

2

Dehydrogenase from Bacillus megaterium IWG3. J. Biol. Chem. 1989, 264, . 9 b. 9)4

3

(36) Filling, C.; Berndt, K. D.; Benach, J.; Knapp, S.; Prozorovski, T.; Nordling, E.; Ladenstein,

4

R.; Jörnvall, H.; Oppermann, U. Critical Residues for Structure and Catalysis in Short-Chain

5

Dehydrogenases/Reductases. J. Biol. Chem. 2002, 277, ).55b ).9 4

6

(37) Schlieben, N. H.; Niefind, K.; Müller, J.; Riebel, B.; Hummel, W.; Schomburg, D. Atomic

7

Resolution Structures of R-Specific Alcohol Dehydrogenase from Lactobacillus brevis Provide

8

the Structural Bases of its Substrate and Cosubstrate Specificity. J. Mol. Biol. 2005, 349,

9

9; b9

4

10


Page 26 of 36

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Scheme 1. Reaction path for the enzymatic oxidation of diols.

3



T215 E214 V198 F197 NADP+ S196 V84 Y164 Y127 K168

S126

4

RMSD ( )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3 2 1 0 0

2000

4000

6000

8000 10000 12000 14000 16000 18000 20000

1 2

Figure 1. Analysis of the substrate binding pockets of WTKpADH using MD simulations. Yellow

3

sticks: NADP+, blue sticks: catalytic triad, green sticks: 1,4-BD, blue spheres: V84, Y127, S196,

4

F197, V198, E214 and T215.

Simulation time /ps

5


Page 28 of 36

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ACS Paragon Plus Environment


1 80

80

60

40

20

50

40

30

60

40

20

20

10

0

0 0

1

2

3

4

Concentration of GBL /mM

(B)100 Yield of D-Phe /%

(A)100 Yield of D-Phe /%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 36

0 0

2

4

6

2 3

Figure 3. Asymmetric preparation of D-Phe with DAADHD94A coupled with NADPH

4

regeneration approaches. (A) Effect of different cofactor regeneration systems. FhG=

5

KpADHV84I/Y127M, (h): KpADHV84I, (h): KpADHY127C, (h): KpADHV84I/Y127C, FiG= WTKpADH,

6

( ): GDH. Reactions were performed with 0.5 mM NADP+, 50 mM PPA, 50 mM 1,4-BD or

7

100 mM glucose, 1.5 kU·L–1 enzymes, and 5 kU·L–1 DAADHD94A. (B) Production of D-Phe and

8

GBL at different PPA and 1,4-BD concentrations employing KpADHV84I/Y127M. FhG & (F):

9

concentrations of D-Phe & GBL at 50 mM PPA and 25 mM 1,4-BD, (

10

Time /h

of D-Phe & GBL at 100 mM PPA and 50 mM 1,4-BD.

11 12


Time /h

)&(

): concentrations

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60




1 2

Table 1. Comparison of the thermodynamic, substrate price and waste amount of various NADPH regeneration approaches. Cosubstrate HCO2H Isopropanol Ethanol H3PO3 Glucose 1,4-BD

3 4

Page 32 of 36

Coproduct CO2 Acetone Acetic acid H3PO4 Gluconic acid GBL

Catalyst FDH ADH ADH/AldDH PDH GDH KpADH

Price [$·mol–1] 1.11 1.29 0.55 3.04 1.21 1.03

T6’ rxn [kcal·mol–1]a –4.6 –6.1 –7.4 –15 –6.9 –8.2

Waste [g·mol–1·prod.] 44 58 30 98 196 43

T6’ rxn of each cosubstrate was cited and calculated according to Ref15, the price of each cosubstrate was cited from commercial source.

a

5


Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1


Table 2. Kinetic parameters of WTKpADH and variants toward NADP+ and 1,4-BD.

Variant WTKpADH

KpADHV84I KpADHY127C KpADHY127M KpADHS196A KpADHF197W KpADHV84I/Y127C KpADHV84I/Y127M KpADHY127C/S196A KpADHS196A/F197W

Kia [WM] a 26.7±4.5 10.7±1.9 757±103 39.7±6.7 44.9±2.8 347±77 493±78 49.7±5.5 194±13 70.8±5.3

KM,NADP+ [WM] 15.8±1.8 30.9±2.2 8.34±2.8 14.7±1.5 24.7±0.9 17.2±2.7 9.99±2.5 3.05±0.7 9.21±1.3 10.0±0.8

KM,1,4-BD [mM] 50.6±6.7 86.6±8.3 21.4±8.6 28.9±4.5 45.9±2.6 12.2±4.5 17.7±5.0 15.1±1.4 31.0±3.3 23.4±1.8

kcat [min–1] 8.76±0.4 80.3±3.9 18.1±0.9 32.4±1.2 13.8±0.2 8.21±0.4 23.0±0.9 30.1±0.4 28.7±0.7 10.2±0.2

kcat/KM to 1,4-BD b [min–1·mM–1]

0.173±0.015 0.927±0.044** 0.846±0.030** 1.12±0.133** 0.301±0.013* 0.673±0.022** 1.30±0.152** 2.00±0.158** 0.926±0.076** 0.436±0.025**

2

aK

3

b

4

statistically significant difference (p < 0.05), ** indicates an extremely significant difference (p < 0.01).

ia

denotes the dissociation constant toward NADP+.

kcat/KM was determined according to kcat and KM,1,4-BD. Statistic analysis was performed: * indicates a

5 6



1

Table 3. Kinetic parameters of WTKpADH and variants toward different diols

Variant

KM kcat kcat/KM KM kcat kcat/KM [mM] [min–1] [min–1·mM–1] [mM] [min–1] [min–1·mM–1]a WTKpADH 74.4±2.2 55.7±1.0 0.75±0.01 18.5±1.5 41.9±0.9 2.27±0.10 ** KpADHV84I 42.5±1.5 51.1±0.8 1.20±0.02 12.3±0.7 64.7±1.7 5.25±0.16** KpADHY127C 63.5±0.1 25.1±0.9 0.39±0.01** 31.8±1.7 36.2±2.2 1.14±0.01** KpADHY127M 14.3±0.9 46.6±1.7 3.27±0.09** 6.50±0.5 39.5±1.9 6.09±0.17** KpADHS196A 75.4±2.3 64.4±1.6 0.86±0.01* 21.3±2.6 46.4±0.8 2.18±0.13* KpADHF197W 61.2±1.9 40.1±1.2 0.66±0.01* 13.5±0.1 26.2±1.2 1.94±0.07** KpADHV84I/Y127C 16.3±0.8 24.8±1.1 1.52±0.01** 12.9±1.3 98.3±2.5 7.63±0.15** KpADHV84I/Y127M 8.20±0.2 19.3±0.9 2.35±0.05** 6.50±1.0 75.2±2.1 11.5±0.15** KpADHY127C/S196A 43.2±0.5 23.7±0.7 0.55±0.01** 21.6±1.0 49.8±1.4 2.31±0.04* KpADHS196A/F197W 50.1±0.5 43.1±1.3 0.86±0.02* 10.7±0.8 35.7±0.7 3.34±0.18** a Statistic analysis was performed: * indicates a statistically significant difference (p < 0.05), ** 2 3 indicates an extremely significant difference (p < 0.01). 4


Page 34 of 36

Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 4. Kinetic parameters of WTKpADH and variants toward NADP+ and 2-HTHF.

1 Variant

WTKpADH

KpADHV84I KpADHY127C KpADHY127M KpADHS196A KpADHF197W KpADHV84I/Y127C KpADHV84I/Y127M KpADHY127C/S196A KpADHS196A/F197W

Kia kW,l a 57.9±6.2 108±23 361±27 129±14 217±34 163±11 961±99 94.0±7.2 288±26 111±7

KM, NADP+ kW,l 64.8±2.6 159±18 107±9 76.8±15.6 456±31 107±5 462±57 80.7±7.7 565±59 100±3

KM, 2-HTHF [mM] 2.23±0.37 1.90±0.96 2.40±0.71 5.90±2.19 1.55±0.65 1.59±0.30 4.70±2.88 4.39±0.92 6.38±2.50 0.85±0.11

kcat [×103 min–1] 2.63±0.09 1.33±0.20 0.464±0.04 2.34±0.39 2.27±0.27 1.09±0.05 0.315±0.07 0.855±0.06 0.374±0.07 0.441±0.01

kcat/KM to HTHF [min–1·mM–1] b 1.18±0.15×103 700±148* 193±40** 397±81** 1.46±0.11×103* 686±77** 67.0±16.0** 195±27** 58.6±12** 519±55**

Fold Change c 6.82×103 755 228 354 4.86×103 1.02×103 51.6 97.8 63.3 1.19×103

2

aK

3 4

b

Statistic analysis was performed: * indicates a statistically significant difference (p < 0.05), ** indicates an extremely significant difference (p < 0.01).

5

c

ia

denotes the dissociation constant toward NADP+.

Fold change denotes the ratio of kcat/KM toward 2-HTHF to kcat/KM toward 1,4-BD.

6




Page 36 of 36

Engineering an Alcohol Dehydrogenase for Balancing Kinetics in

Recommend Documents